How Brands Win in AI Search

Insights from Campaign's latest webinar

Credit: Freepik

The Shift

I recently attended Campaign's latest webinar, How Brands Win in AI Search: From Clicks to Consideration, hosted in partnership with digital agency Green Park Digital. The panel brought together senior voices from Diageo, L'Oréal-backed beauty start-up Noli, and Green Park's own search and insights teams. It was a very insightful conversation filled with practical advice that you will find sprinkled throughout this blog.

If you work in marketing or PR, here is the fundamental truth that underpinned everything discussed: you are no longer competing for clicks. You are competing for the narrative. That single mindset shift challenges almost every assumption that brands have built their SEO strategies on for the last decade, and it is the lens through which everything else in this piece should be read.

This should also be music to the ears of PR & Marketeers. For a long time, we have often struggled to get across the value of our work. Now, as markers such as Share of Voice, Sentiment, and Earned Media come to the fore, we can point to the direct impact our work is having in helping brands show up and stand out in the world of AI.

Traditional SEO vs AI Search

Traditional SEO was, in many ways, transactional. You targeted a keyword, you ranked for it, and traffic followed.

AI search works differently. Large language models (LLMs) like ChatGPT, Google's AI Overviews, and others are not returning a ranked list of pages. Instead, they are forming an opinion about your brand by synthesising signals from across the entire internet and presenting a single, authoritative answer. They decide what to trust, how to describe you, and what role your brand plays in a given category. What this means is that brands have to take a much more holistic approach when it comes to optimising for AI Search to match the holistic approach LLMs take in deciding what information to include in their answers.

Why all the bother? I hear you ask. Well, the stakes are significant. Direct-to-consumer (DTC) brands and publishers are already seeing referral traffic decline by 30 to 50 per cent or more. Yet there is a compelling silver lining: traffic that does come through from AI-assisted search tends to be far higher-intent. Research by Semrush has found that an AI search visitor is 4.4 times as valuable as the average visit from traditional organic search, based on conversion rate (source: Semrush, July 2025). Pretty lucrative then, eh? But how do you, as a brand, capture this high-intent audience? Let's dive into it...

The Four C's: A Framework for AI Visibility

Green Park Digital's Head of Omnichannel Search and Insights, Sam Barker, shared a framework developed through conducting AI visibility audits across a wide range of brands. The 'Four C's' offers a practical lens through which brands can assess and improve their standing in LLM-generated answers. They are:

1. Core: Your Owned Ecosystem

Your website and content are the single source of truth. But most brand sites are too focused on aesthetics and under-engineered for the conversational, FAQ-style content that LLMs retrieve.

2. Third-Party Endorsement

Media coverage, expert commentary, customer reviews, and industry recognition. LLMs cross-reference your self-description against external signals. If only you are validating your own claims, you're at a disadvantage.

3. Online Conversations

Your presence on Reddit, TikTok, social forums, and review platforms. This content feeds directly into LLM training data — a critical signal that most brands are not yet actively managing.

4. Product-Level Data

For DTC and e-commerce brands, this means product detail pages (PDPs): consistent product information, pricing, and availability across all retailers. An AI will only recommend a product it sees it can trust.

The key insight from the Four C's is that consistency across any of these four areas is a must. If your brand website says one thing, a third-party review says another, and your Reddit presence is non-existent, an LLM will not attempt to reconcile the contradiction. It will move on to a competitor with a more coherent story.

How to AI-proof your content strategy

Credit: Freepik

Firdaous El Honsali, CMO at Noli, offered a complementary framework drawn from her experience in the beauty industry which provides three fundamental pillars brands should use as the foundations of their AI content strategy. The mistake, she argued, is that marketers rush to the glamorous third pillar while neglecting the less glamorous foundations that make it work.

Pillar One: SEO Remains the Backbone

AI hasn't killed traditional SEO and many of the key aspects of it are still vital. LLMs need signals of well-structured, clearly organised information. Brands need a robust semantic architecture that both Google and LLMs can parse easily: proper FAQs, clear product content, and pages that genuinely answer real consumer queries. The critical shift is from optimising for keywords to optimising for intent. The content you create needs to serve Google and LLMs alike, reflecting the way people now talk to AI in longer, more conversational queries rather than short keyword searches.

Pillar Two: Consistency Across Every Touchpoint

LLMs reward brands that show up consistently with the same narrative everywhere. This is both a content challenge and an organisational one. Whilst your brand marketing, comms, PR, SEO, and influencer teams may operate independently, its vital their work is underpinned by a unified strategy. If not, AI search will expose those misalignments very quickly. Brands need to define the topics and attributes they want to own, then build a coherent body of expert content (e.g. blog posts, white papers, research, and expert partnerships) that reinforce those themes across every channel.

Pillar Three: Authority and Third-Party Recommendation

This is where PR has a genuine and significant role to play. Getting your brand's story into authoritative external sources such as top-tier publications, Wikipedia, expert commentary, industry bodies, and customer reviews is what signals to an LLM that it can trust you. This is, in many ways, the comeback of traditional PR. The brands that have maintained strong relationships with credible media and thought leadership platforms are quietly well-positioned.

However, authority built on an inconsistent or poorly structured foundation is wasted effort. You cannot shortcut to pillar three if pillars one and two are not in place.

What to Measure Instead of Traffic

Credit: Freepik

Near the end of the webinar, someone asked a good question about what metrics should be top of the list when it comes to measuring your efforts to improve your AI visibility. Here are the key metrics Sam Barker put forward:

• Share of Voice: Are you being mentioned when relevant prompts are entered into LLMs? Benchmark this now, before you begin any optimisation work, so you have a baseline from which to measure progress.

• Sentiment: For the first time, brands can see how an LLM describes them. ChatGPT and similar tools offer a window into brand perception that has never previously been available through search. Negative sentiment is now something you can more easily identify, track, and address.

• Messaging Consistency: Is the narrative the LLM is surfacing aligned with your brand strategy and what you want to be known for? There may be a gap between how you describe yourself and how third parties describe you — and that gap is where you are losing ground.

• Baseline Benchmarking: Capture your current share of voice and sentiment before starting any AI visibility work. Without this, it is extremely difficult to demonstrate progress and impact.

As Sam put it: "We've never had access to this before. We couldn't see what Google thought about your brand. We can now see what the perception is in ChatGPT." That is a remarkable new capability and one that brands should be making the most of.

The Role of Reddit and Social Search

Interestingly, in an audience poll which was conducted during the webinar, social search ranked last as an area of brand concern - just 2 per cent of attendees flagged it. The panel all agreed that this is a significant blind spot.

Reddit in particular deserves serious attention from marketing and PR professionals, for two distinct reasons.

First, it is an extraordinarily rich source of unfiltered consumer insight. The conversations happening on Reddit represent a level of authenticity that no survey or focus group can replicate. From the sharing of personal stories, product comparisons, and category debates, brands monitoring these conversations can get a glimpse into what consumers think of their products, as well as adjacent topics and concerns they also need to build credibility in.

As Firdaus noted, Reddit is "an incredible mine for new innovations" and can reveal the kind of real, unvarnished consumer voice that can shape product development as much as marketing strategy.

Second, and critically for AI search: Reddit is one of the most heavily weighted sources in LLM training data. When an LLM forms its view of your brand, it is drawing on exactly the kind of real, peer-to-peer conversations that Reddit hosts at scale. A brand that is absent from, or poorly represented in, those conversations is ceding ground it may not even be aware it is losing.

Sue Jones, Chief Digital Officer at Diageo, reinforced this by highlighting ratings and reviews as another heavily prioritised signal: "We're seeing [ratings and reviews] as being incredibly popular in this conversation. It it seems like that is what the LLMs are prioritising, what real actual people are saying."

This makes reputation management and proactive review strategy a far more commercially significant activity than it might previously have appeared.

The same logic applies more broadly to TikTok, YouTube comments, and other social forums. The panel was careful not to suggest brands should manufacture Reddit presence, however. When it comes to brands on social media, and especially on Reddit, authenticity matters. Plus, LLMs are good at detecting thin or incoherent signals. Rather, brands should be monitoring these conversations, genuinely engaging where appropriate, and ensuring the real consumer sentiment around their products is something they understand and can respond to.

Practical First Steps

There is no quick and easy way to drastically boost your AI visibility; however, the panel had some good advice for easy, quick wins that brands should be implementing first to ensure that your AI visiblity strategy is built on solid foundations.

Here they are:

1. Run an AI visibility audit: Before you change anything, understand where you stand. What is your current share of voice in LLM responses? How are you being described? Where do you appear and where are you absent? Green Park's Optic audit framework offers one approach, but even a manual audit — entering relevant prompts into ChatGPT and Google's AI Overviews — will reveal useful gaps.

   
   

2. Identify your biggest lever: Is your biggest gap in earned media, technical SEO, outdated content, or product page readiness? Prioritise ruthlessly. As Sam Barker put it: "You need to see where your biggest opportunity is and go after that first and...do a great job of it [rather than]...trying to boil the ocean"

   
   

3. Define your narrative: Agree internally on the topics, attributes, and stories your brand wants to own and make sure every team (brand, comms, SEO, influencer) is aligned to that same strategic framework.

   
   

4. Invest in third-party credibility: Even modest investment in earned media, i.e. getting your story into authoritative publications, building expert endorsements, and encouraging customer reviews, can meaningfully shift how LLMs perceive and represent your brand.

   
   

5. Start monitoring Reddit and social signals: Set up regular monitoring of your brand and category on Reddit and other forums. Treat this as both an intelligence tool and an indicator of how LLMs are likely to represent you.

   
   

6. Agree your new metrics: The measurement conversation is one of the most important (and often most difficult) internal challenges. Agreeing on share of voice and sentiment as primary KPIs, alongside revenue and commercial outcomes, is essential to sustaining the strategy over time and to ensure you can accurately monitor the impact of the work you're doing.

Conclusion

With the world of AI evolving at such a rapid rate, it can be tempting to adopt an attitude of simply waiting to see where the dust settles. Whilst it's true that we're seeing AI continue to expand at an incredible rate (especially when it comes to paid AI search ads, which are very limited currently but highly likely to expand), the brands that will win in AI Search are those that start building strong foundations now.

Links to Speakers

I'd like to also take this opportunity to thank Campaign for putting the webinar on. You can find social links to those involved in the discussion here:

Lucy Shelley - Tech Editor, Campaign

Sue Jones - Chief Digital Officer, Diageo

Chris Pearce - Managing Director EMEA & US, Greenpark

Sam Barker - Head of Omnichannel Search & Insights EMEA & US, Greenpark

Firdaous El Honsali - Chief Marketing Officer, Noli